On Citizenfour: A Conversation with Bernard Harcourt

Photo by Laura Poitras / Praxis Films, licensed under CC BY SA.

Laura Poitras’s 2014 documentary film Citizenfour portrays Edward Snowden and the whistleblowing operation that exposed surveillance practices within the National Security Agency (NSA) ­­­­and other signal intelligence agencies worldwide. The documentary, which won the 2015 Oscar for Documentary Feature Film, raises timely questions about surveillance and the limits of privacy in the age of Big Data. Poitras’s previous documentary films—My Country, My Country (2006) and The Oath (2010)—meditate on U.S.–Middle East relations in the post-9/11 era. All three films document individual actions that are at odds with the values espoused in their immediate context; thus, at the center of Poitras’s work are ethical questions of dissent, law, and patriotism.

This interview with Bernard Harcourt, the Isidor and Seville Sulzbacher Professor of Law and Director of the Center for Contemporary Critical Thought at Columbia University, was conducted as part of “Spectacle and Surveillance,” a seminar cotaught in 2015 by Harcourt and W. J. T. Mitchell, Professor of English and Art History at the University of Chicago. The purpose of the seminar was to “reflect on the dialectical pairing of spectacle and surveillance as modes of image power—that is, power over subjects in the case of spectacle, over objects in the case of surveillance—and as modes of governing in our contemporary age of Big Data.” Harcourt’s most recent book is entitled Exposed: Desire and Disobedience in the Digital Age (Harvard University Press, 2015).

Andrés García Molina: Let’s start by talking about Edward Snowden. We could say that his actions are a form of dissent. However, such dissent is one that is not only a refusal to be complicit, but also one that involves an ethical disposition towards disseminating information. Snowden not only resigned from his position, but also felt that it was his responsibility to reveal information to a broad public. What kind of dissent is Snowden’s, in that it has this component of dissemination?

Bernard Harcourt: I am not sure that dissent is the right word, or that it captures fully what Snowden is engaged in. I think it would be more like disobedience than dissent. I say that because he is breaking the law—it is fair to say he is breaking the law. He is violating legal rules about the classification of information that are established by American law. Dissent can have many different meanings, but dissent can be merely verbal disagreement; it can be disagreement with a decision. In legal parlance we talk about dissenting opinions written by judges who disagree with the results of a case. One can push dissent into acts of explicit disobedience, but that is not always the case. Since dissent is such a broad term, I think it would be better to use a different one for Snowden, and I think that the right term for Snowden would be disobedience.

AGM: What would you say about the additional component I mentioned, of having an ethical disposition toward disseminating information? Ultimately, Snowden’s objective was to expose a specific set of activities in which various government agencies are engaged. In that sense, how do Snowden’s acts compare, for example, with Julian Assange’s?

BH: Here is one difference between someone like Snowden and someone like Assange. Snowden’s act of disobedience is an act that could be justified on the grounds of necessity: it is necessary for him to break the law and to publicize the kinds of surveillance that are going on in order to promote a set of interests shared by Americans. So he might invoke the notion of necessity or the lesser of two evils, a conventional defense in criminal law. Significantly, this assumes that there is consensus over the overarching values of society and that an individual’s actions are actually necessary to promote those values in society. I do not think that Snowden has any objection to conventional American liberal values, and his actions were not intended to undermine those values, nor to undermine national security, nor even to undermine, say, the United States’ involvement in Iraq or the Middle East. His actions do not necessarily imply an overarching political agenda. They are about surveillance and privacy and what the government can know, is able to know, should or should not know, and about how that relates to traditional civil rights protections and to traditional protections about privacy.

Assange’s case provides an interesting point of comparison. Assange would not defend WikiLeaks in terms of necessity, as in the upholding of generally shared values. His case is possibly not one of disobedience because he is not constrained by the classifications of law in the United States, for example. We could say that he is not engaging in legal disobedience, because his is not an intervention that is necessary to uphold “our” civil rights. It’s a political intervention, a political strategy that is geared to a struggle over power.

AGM: Snowden’s actions remind us that a surveillance apparatus is ultimately predicated on the combined actions of individuals who might be doing something as mundane as operating a computer, entering information in a database, or installing hardware. These are individuals who may or may not comply, may or may not engage in whistleblowing. So, in thinking about Snowden’s specific case, I wanted to ask you about an apparent contradiction in Citizenfour. Snowden openly says he is not the story and he does not want to be at the center of it. He is concerned that the media will try to distract from the information he is revealing by focusing instead on him as an individual, ultimately attempting to discredit the significance and implications of his actions. He explicitly resists that type of attention, but in the end the film is very much about him. Citizenfour reveals another side of the story as we get a sense of who Snowden is and what he believes in. What do you make of this paradox?

BH: I think it is impossible to extricate the individual from these revelations. As a public, we care about the authenticity and completeness of the revelations. One way in which we interrogate both authenticity and completeness is through a notion of motivation, in this case, Snowden’s. Why is it that he reveals this information? Ultimately, it would be unwise to rely too strictly on a subjective interpretation of Snowden’s motivations, because motivation and intent are extraordinarily difficult notions to grasp and determine. There, we might be going down a rabbit hole. However, I think we tend to judge the revelations and what they tell us about American surveillance practices based in part on the motivations of the person who is revealing the information. If we thought, for instance, that Snowden’s real motivation was to help a foreign government or that his real motivations were purely financial, then we would question whether he is providing a full archive or instead presenting a selective archive that would be particularly remunerative or damaging. So, in part, we turn to motivation to be able to then make certain judgments about whether we think that we have obtained the full and complete and reliable view of what the NSA is doing. Again, relying too strictly on a notion of motivation is probably a mistake, but I think that it is inevitable.

And to a certain extent I think that Laura Poitras and Glenn Greenwald were keenly aware of this, and that might be why they framed Snowden’s presentation of self largely through the lens of motivation. He has the ten-minute presentation of self on that fateful Sunday, right after a few of the articles have gone public, and a lot of it is dedicated to addressing questions like: “Why did you do this?” “What is your ambition?” “Were you simply trying to help the Chinese?” And it’s interesting because when he is asked those questions, he does not really answer them very well at first. He goes off on a bit of a dissertation about the Chinese government. So Glenn Greenwald has to ask him again. I think that is one of the final questions; Greenwald stops him and asks him explicitly if he was engaged in espionage for a foreign government. And then Snowden responds again. That is, unquestionably, a part of the documentary. It is a subtext about the genuine nature of his motivation.

AGM: In your own work, you have discussed the idea of an expository society, in which individuals willingly provide information and intimate details about their personal lives to different companies, whether on or offline. We willingly use services that we know, to a degree, imply giving up our privacy. I wonder, however, about the degree to which we can assume that most people actually know that they are giving up their privacy. Can we talk about an expository society without assuming that people know they are exposing themselves? And if there is no open consent, can we still call those forms of usage expository?

BH: We might say that it is not so much willingly, but knowingly. I do not think that it is necessarily people’s purpose for their information to be available to the NSA, but I think that there is—or there should be—enough knowledge now, predominantly as a result of the Snowden revelations, that these materials are then retrievable by search engines and by signal intelligence agencies. There are at least two elements to the expository nature of the expository society. The first is the one that you are focusing on right now, which has to do with the fact that we are exposed and that people can look at us. But the other is the way in which we are presenting ourselves, not for surveillance, but for other people. So there is a dimension of this argument that is about self-presentation on Facebook or other social media, about crafting a profile, a personality. There is an expository dimension to this behavior, even within the small expository realm of the friends that one has on Facebook, Instagram, Twitter, and so forth. There is a publicness and a dissemination that did not exist before, an expository element that exists regardless of our knowledge of the NSA and other signal intelligence agencies.

There are several dimensions to the expository society. One of them, we do engage in knowingly and willingly, and that is to our friends and followers on social media. One of them is to the media and the technologies themselves, which we are aware of, because of all the targeted advertisements that come our way. It is impossible for you to do a search on Google and not know that you have been tracked when those same searches, those same products, come back to you later on. Still another dimension is to signal intelligence agencies and other people who are engaging in surveillance, and there, too, it is with knowledge. We know it. If we do not know it, it is that we are willfully ignoring the newspaper reports about it. But all of those dimensions together amount to something that could be called an expository society.

AGM: So, along similar lines, you have discussed a “data double,” or a digital self, an idea that Kevin Haggerty and Richard Ericson (2000) also engage: the aggregation of . . .

BH: Our digital traces. An aggregation of our digital traces, which ends up constituting a self that is not the analog self, but just the traces that we have left everywhere and that can be tied together to create, practically speaking, a personality.

AGM: I will ask a very basic question, then. In what way does this differ from older styles of aggregation such that, say, a state knows when and where you were born, your blood type, who your mother is, your father, your siblings? The state might know where your family lives, their tax and employment history, and so forth. So, what is new, what is different now?

BH: It goes far deeper into your . . . I think what we might want to call your desires, because this aggregation contains all of your consumption and all of your purchases and entertainment and readings and pastimes and pleasures.

AGM: So, pervasiveness?

BH: It is a difference in kind, because of the quantity of information that we have. It is also about every minor disparity in the amount of time one spends on different websites. You are getting such a rich profile of a person’s curiosities and interests and investments and emotional life and tastes and desires.

AGM: Yes, and then there is the potential searchability of all of that. Let me go back to Citizenfour. At the end of the film, we see Glenn Greenwald’s hand tearing a sheet of paper. The camera zooms in on the bits of paper, and we see a hand picking them up and scrambling them. How do you read that last moment?

BH: Well, in a number of ways. First, it suggests that there is another informant, or another person that is providing information. But, second, it is an act of shredding, an old-fashioned act of shredding that does not use the office technology. It is an old-fashioned way of trying to keep something secret. You can tear paper into little pieces and then you can mix them up. There’s something about the movement, too, about the hand motion. There’s something sensual about disrupting knowledge that is reflected in that last scene.

AGM: It is a sort of inverse operation of digital aggregation.

BH: Yes, a dispersal of all the little bits and then shaking them up so you cannot put them together again.

AGM: Despite all the media coverage of Snowden’s collaboration with Poitras and Greenwald, despite Citizenfour winning an Oscar, the outcome of it all could be said to be anticlimactic. It is hard to say that Snowden’s revelations have resulted in changes in policy or in a demonstrably more knowledgeable public. No revolution has been unleashed. How do you make sense of this gap between revelation and reaction?

BH: Recent surveys have shown that roughly three-quarters of Internet users in the United States are aware of Edward Snowden and of his revelations concerning the NSA. They indicate that approximately 36 percent of those users have done something to change the way in which they use the internet, or to change the way they understand their privacy in terms of how they might protect themselves while online. So the question might be whether we think that is a large impact, or whether the news is falling through the cracks. Now, of course, it is only 75 percent—76 percent, to be precise—of Internet users and not 75 percent of respondents. I find that a little bit unsettling, insofar as I would expect practically all Internet users to be aware of the Snowden revelations. It is also interesting to see that other countries have higher levels of awareness, including Germany, China, Brazil, Sweden, and Hong Kong.

Now, one could say the revelations had a significant effect on public debate during the summer of 2013, and maybe for the following year, but I do think that there is a cycle to the news and that we will gradually lose focus on the issues that Snowden raised and return to a state of normalcy. That makes it even more important for those of us who are concerned about this to continue to raise these issues, to write about them and to keep them at the forefront of public consciousness. That requires a lot of work, just to maintain awareness and consciousness. It is not even reform or enacting changes. There may be some changes to the 215 program, which is the collection of telephony metadata in bulk. What I foresee is that information will be required to be held by the telecoms rather than turned over to the NSA. And depending on what you think the threat is, that is not really much of a solution. If you view the problem as the amalgamation of security with telecom and tech companies, if you see surveillance as a problem that extends to social media and retailers as well as the NSA, then you are not going to think that having the telecoms hold the metadata makes much difference. I expect that this is about the extent of legislative reform that is going to take place, and in my opinion, it is a red herring. It does not address the problem at hand. [Editor's note: Shortly after this interview was conducted, the USA Freedom Act was signed into law; under it, the NSA no longer has direct access to telephony data without prior authorization from a United States Foreign Intelligence Surveillance Act (FISA) court. Telecommunications companies continue to collect and store this data. Since the legislation only regulates access to telephony data, the NSA can still access massive data originating from the Internet and social media.]

AGM: Is the implication that the metadata will still be held so that it can be summoned when needed or deemed convenient?

BH: Right. The issue is that it will be effectively held by the telecoms, and actually they are probably going to be compensated for this. So it is an economic win-win situation for the telecoms, but it does not mean that the information is not readily accessible for intelligence services and also by other entities.

AGM: Throughout Citizenfour, we see the exteriors of buildings, both faraway shots of NSA locations and also the outer façades of hotels and commercial buildings. To me, these shots suggest an ongoing theme of impenetrability and opacity. The film’s form, too, reflects it: it’s thriller-like in the sense that it does not reveal the entire story from the beginning, but rather keeps the viewer in suspense wondering, first, who Snowden is, and then, what will happen to him. Opacity might also describe the way in which agencies like the NSA operate behind curtains. It might also describe encryption techniques that people like Snowden employ as a defense mechanism, an act of countersurveillance. Perhaps the funniest moment of the film comes when Snowden uses his “hiding cloak”—a towel wrapped around his head and torso—to enter a password into Greenwald’s computer. Greenwald then enters his own password, which is met with a sharp retort from Snowden: “So your password is, like, four characters long?” A broad question, then: can you comment on these techniques of opacity?

BH: This is a good question, the question of opacity. I think that is one of the most interesting theoretical issues, the relationship between visibility or transparency and opacity. Frank Pasquale (2015), of course, has written nicely about this in his book The Black Box Society. I tend to think about it in terms of what I call virtual transparency, which has an element of opacity within it. A lot of our information is easily accessible and practically transparent, although some of it is going to be distorted because some of it is false and some of it is going to be intentionally falsified. At the same time, many, if not most, of the technologies we use are themselves opaque, like recommender technologies and algorithms themselves. A lot of the entities that are watching us remain opaque. The NSA, for instance, tries to cultivate a form of opacity in an ambivalent way. At times, the NSA wants everyone to know of its extraordinary capabilities, in order to deter what it considers to be security risks. At other times, it wants us to fall into a certain form of complacency and forgetfulness in order for us to expose ourselves more.

This relationship to opacity is rich and complex. Both transparency and opacity turn out to be ambiguous. What one gets is a somewhat distorted view of the world that has a lot of reflections, a lot of information that gets mirrored through virtual transparence even as some hidden areas and dark areas remain. Efforts at cloaking leave us with something that can be imagined, to use an architectural example, as a kind of glass and steel mesh with reflections, transparency, and mirroring.

AGM: To bring it back to Citizenfour, though: that is such a charged moment, when Snowden makes his joke about the magic towel and reflects slyly on passwords. That moment reminds us that opacity is a type of technique that anyone can wield. In the end, opacity is not reserved for the NSA.

BH: Right. Snowden easily discerns that Greenwald has a very short password, and he suggests that it would be easy to decrypt through algorithmic processes. So, all of those forms of creating opacity may also be subject to decryption and transparence, as Snowden himself revealed about the NSA. There is a constant, shifting relationship between transparency and opacity that seems to constitute our present age.


Haggerty, Kevin D., and Richard V. Ericson. 2000. “The Surveillant Assemblage.” British Journal of Sociology 51, no. 4: 605–22.

Pasquale, Frank. 2015. The Black Box Society: The Secret Algorithms That Control Money and Information. Cambridge, Mass.: Harvard University Press.

Related Readings

Deleuze, Gilles. 1990. “Postscript on Control Societies.” October, no. 59: 3–7.

Foucault, Michel. 1977. Discipline and Punish: The Birth of the Prison. Translated by Alan Sheridan. New York: Vintage.

Greenwald, Glenn. 2014. No Place to Hide: Edward Snowden, the NSA, and the U.S. Surveillance State. New York: Metropolitan Books.

Monahan, Torin. 2011. “Surveillance as Cultural Practice.” Sociological Quarterly 52, no. 4: 495–508.

Stallabrass, Julian. 2011. “Negative Dialectics in the Google Era: A Conversation with Trevor Paglen.” October, no. 138: 3–14.

Vaidhyanathan, Siva. 2011. The Googlization of Everything (And Why We Should Worry). Berkeley: University of California Press.

Webster, James G. 2014. The Marketplace of Attention: How Audiences Take Shape in a Digital Age. Cambridge, Mass.: MIT Press.